Authors

BMJ Open Science uses open peer review. This means that the reviewers will know who wrote the article and the reviewers’ identities will be disclosed to you when an editorial decision is made. We also publish the peer review comments alongside any article we publish under the same licence (CC BY).

Read more information about reviewing for BMJ Open Science.

Plagiarism is the appropriation of the language, ideas or thoughts of another without crediting their true source and representation of them as one’s own original work. BMJ is a member of CrossCheck by CrossRef and iThenticate. iThenticate is a plagiarism screening service that verifies the originality of content submitted before publication. BMJ runs manuscripts through iThenticate during the peer review process. Authors, researchers and freelancers can also use iThenticate to screen their work before submission by visiting www.ithenticate.com.

We have a strict, forward-thinking open data policy. This means that the data underlying the findings in your article should be made publicly available no later than when the article is published.

Data files may not be uploaded as supplementary material. Data hidden in supplementary files are not easily discoverable, do not have unique permanent identifiers, and are not protected in perpetuity, unlike the article.

In exceptional circumstances, where there are restrictions on the availability of data, the Editor may agree to publish an article without publicly available data so long as the location, availability, and reason for restriction of the data is made clear in the article. A request for an exception must be made at submission or by contacting our editorial office in advance if you think you cannot comply with our policy.

The journal follows the TOP (Transparency and Openness Promotion) guidelines and promotes the FAIR data principles.

Read more about our data sharing policies.

BMJ Open Science mandates ORCID iDs for the submitting author at the time of article submission; co-authors and reviewers are strongly encouraged to also connect their ScholarOne accounts to ORCID. We strongly believe that the increased use and integration of ORCID iDs will be beneficial for the whole research community.

Please find more information about ORCID and BMJ’s policy on the BMJ Author Hub.

BMJ Open Science encourages a number of initiatives to help the work that we publish be reproducible.

Penelope pre-submission manuscript checking

Before submitting your manuscript, consider checking it with Penelope, an award-winning online tool that checks the completeness of scientific manuscripts and gives immediate feedback to authors. It has been customised to BMJ Open Science guidelines to help you prepare for submission.

Reporting guidelines

As part of our commitment to reproducibility, we encourage authors to upload relevant reporting checklists as files for editors. You can find appropriate checklists, such as ARRIVE, at the EQUATOR and MERIDIAN sites.

We particularly encourage compliance with the Landis 4 criteria: randomisation, blinding, sample size calculation, and inclusion/exclusion criteria (doi:10.1038/nature11556), where appropriate. We are keen to be transparent about any limitations of methods used and non-compliance will not necessarily determine the article’s outcome.

When submitting an article you will be asked about the following:

Randomisation

Please describe how samples/animals were allocated to experimental groups. If randomisation was not described please explain why this was not possible or was not appropriate.

Blinding

Please describe whether investigators were blinded to group allocation during the experiment and/or analysis. If blinding was not performed please explain why it was not possible or was not appropriate.

Sample size calculation

Please describe how sample size was determined. If a sample size calculation was not performed please explain why it was not possible or was not appropriate.

Inclusion/exclusion

Please describe any a priori inclusion/exclusion criteria and any data exclusions with reasons.

A rapid response is a moderated but not peer reviewed online response to a published article in BMJ Open Science; it will not receive a DOI and will not be indexed. Find out more about responses and how to submit a response.

Original research

At BMJ Open Science, we divide our research articles into three types: confirmation studies, exploratory studies, and meta-research. This distinction values the differences in study designs, threats to validity, and the inferences specific to each research type. Further reading about how we came to these article types is in Kimmelman et al, 2014, PLOS Biology (doi:10.1371/journal.pbio.1001863) and Mogil et al, 2017, Nature (doi:10.1038/542409a). Each article type is described in full below.

All research articles must have a clear and justified research question and follow an IMRAD format (Introduction, Methods, Results and Discussion).

Please include in your article:

• Title page: include authors and affiliations including ORCID identifiers.
• A structured abstract (max. 250 words) including objectives, methods, results and conclusions
• Please include a ‘Strengths and limitations of this study’ section after the abstract. This section should be no more than 5 bullet points relating specifically to the methods – not the results of the study. This will be published as a summary box after the abstract in the final published article.
• Introduction: a clear statement of the main study aims and major hypothesis/research question
• Methods: details of the design, the subjects included, the Research Resource Identifiers (RRIDs) for critical reagents and tools, the main methods employed, and clearly defined primary/secondary outcomes (where appropriate). Please also include your protocols.io DOIs if appropriate.
• Results: the main findings with 95% confidence intervals, if appropriate.
• Conclusions: primary conclusions and implications. Do not go beyond the data.
• Full references
• Data availability statement
• Authors’ contributions as per the CRediT taxonomy.
• Funding statement
• Competing interests statement

Confirmation studies 

Confirmation studies are hypothesis-testing preclinical research studies. In addition to a clear and justified research question, confirmatory studies should be conducted and reported to the highest level of rigour.

Confirmatory primary research studies have two features:

1. • The study must have been pre-registered in a protocol format, a registered report (see below for further details) or in a registry. This must be publicly available.
   • They must adhere to the highest levels of rigour in design, analysis and reporting. Where
     appropriate, studies must be randomised, blinded, state how they decided the number of animals included to ensure power, and describe exclusions. We encourage authors to refer to the Experimental Design Assistant for guidance with experimental design. Reporting must comply with the appropriate reporting guidelines for the study type (e.g. you may use ARRIVE guidelines for in vivo studies).

Exploratory studies

Exploratory studies (hypothesis-generating) are exploratory investigations that seek to generate robust pathophysiological theories of disease or intervention efficacy. These studies may be published alongside confirmation studies but must clearly be delineated as exploratory or submitted as stand-alone studies. The design, analysis and reporting of these studies will be held to the same high standard as confirmation studies but typically are not pre-registered, are not required to present a priori power calculations and do not necessarily use well-established techniques. Where no a priori power calculations have been provided, inferential statistics (e.g. p-values) are not appropriate and should not be presented; however, descriptive statistics (e.g. averages ± variances/confidence intervals) are appropriate and encouraged.

Meta-research

Meta-research studies investigate research itself and provide an evidence base for improvement and/or decision-making, including systematic reviews. They must address research questions relevant to preclinical research. We require pre-registration (e.g. via PROSPERO, OSF) or publication of a protocol prior to submission of the final article. Pre-registration or protocols must be publicly available and completed before completion of data collection.

BMJ Open Science will consider other types of primary research that is directly relevant to the design, conduct, reporting, or publishing of preclinical research.

Registered reports are a form of empirical article in which the methods and proposed analyses are pre-registered and reviewed prior to research being conducted. Data collection must not have begun.  Read more about the guidelines for submitting a Registered Report. Authors can also submit a Registered Report from the Peer Community in Registered Reports initiative.

Please include a ‘Strengths and limitations of this study’ section after the abstract. This section should be no more than 5 bullet points relating specifically to the methods – not the results of the study. This will be published as a summary box after the abstract in the final published article.

BMJ Open Science will consider protocols for any study design, including observational studies and systematic reviews. Protocol manuscripts should report planned or ongoing research studies. If data collection is complete, we will not consider the manuscript. For protocols of preclinical systematic reviews, we encourage reference to de Vries et al. (2015). We strongly encourage you to register your protocol. We recommend the Open Science Framework, or Prospero for systematic reviews. Protocols for studies that require ethical approval, such as some in vivo studies, will not be considered without having received that approval.

Reviewers will be instructed to review for clarity and sufficient detail. The intention of peer review is not to alter the study design. Reviewers will be instructed to check that the study is scientifically credible and ethically sound in its scope and methods, and that there is sufficient detail to instil confidence that the study will be conducted and analysed appropriately.

• Title: should include the word “protocol” or clearly reflect that the article is reporting a protocol.
• Abstract: this should be structured with the following sections. Introduction; Methods and analysis; Ethics and dissemination. Registration details should be included as a final section, if appropriate. Abstracts for meta-research should be structured as follows: Objective (explain why you are doing the research); Search strategy (types of study to be included and which sources will be searched); Screening and annotation; Data management and reporting.
• Please include a ‘Strengths and limitations of this study’ section after the abstract. This section should be no more than 5 bullet points relating specifically to the methods – not the results of the study. This will be published as a summary box after the abstract in the final published article.

BMJ Open Science will consider primary research studies that describe novel methods or significant improvement in standard research techniques of preclinical research close to medicine. Methods papers should provide researchers with new tools required to improve the validity of their research. Methods papers have the same IMRAD format, and the design, analysis and reporting of these studies will be held to the same high standard as original research articles. Step-by-step protocols required to allow replication of the methods must be deposited in a public and permanently available community-recognised repository (e.g. Dryad, Figshare).

Please include a ‘Strengths and limitations of this study’ section after the abstract. This section should be no more than 5 bullet points relating specifically to the methods – not the results of the study. This will be published as a summary box after the abstract in the final published article.

BMJ Open Science will consider data descriptor articles of preclinical studies or studies relevant to preclinical research. These articles should describe scientific data to facilitate data-sharing and reuse; their focus is to enable others to reuse data rather than presenting new hypotheses, analyses or interpretations. Descriptor articles combine traditional narrative content with curated structured metadata. Data descriptor articles should include detailed descriptions of the methods used to collect the data and technical analyses to support the quality of data acquisition. Peer review evaluates the rigour with which experiments were conducted during data acquisition. Data must be stored in public and permanently available community-recognised repositories (e.g. Dryad, Figshare).

• Data descriptor articles should include the following:
• Title page: include authors and affiliations
• Abstract (max. 250 words)
• Introduction and summary
• Methods: details of the design, the subjects included, the main methods employed and any inclusion/exclusion criteria
• Data records
• Technical validation: include descriptive statistics
• Usage notes: describe any code or software required for data reuse
• Full references
• Authors’ contributions
• Funding statement
• Competing interests statement

Please include a ‘Strengths and limitations of this study’ section after the abstract. This section should be no more than 5 bullet points relating specifically to the methods – not the results of the study. This will be published as a summary box after the abstract in the final published article.

Reporting and conduct guidelines within scope will be considered. We encourage reporting guidelines to be developed in accordance with Moher et al. (2010).

These articles are usually commissioned or invited; however, authors are invited to discuss possible topics for submission directly with the Editor.